Abstract

N. Deepa Devi* and S. Mariappan

Genetic diversity among 50 genotypes was worked out using Mahalanobis D2 statistic. Based on D analysis, the genotypes were grouped into four different clusters, where the cluster I possessed higher number (32) of genotypes followed by the cluster II (2), III (15) and IV (1). Clustering pattern revealed that geographical diversity was not associated with genetic diversity that is, genotypes collected from same location were grouped into different clusters. The maximum inter-cluster distance was observed between the clusters III and cluster IV and that of minimum in between the Clusters I and Cluster II. In case of intra-cluster distance, the maximum distance was observed in the cluster III and that of minimum was observed in the cluster IV. Among the nine traits studied, maximum contribution was made by average fruit weight (31.84%) followed by days to first male flower appearance (25.96%), yield per hectare (23.59%) and fruit length (17.31%). Considering cluster mean, the genotypes of cluster I and cluster III could be selected for yield and yield attributing characters. The wider genetic diversity was observed in cluster II, III and IV which indicate the potentiality of this diverse genotype collection for providing basic material for future breeding programmes.